What problem does it solve? Writing agent skills without testing them is guesswork: you cannot know whether a SKILL.md actually changes agent behavior until you watch an agent fail without it. This Skill applies the RED-GREEN-REFACTOR cycle to skill authoring so every skill is verified against real baseline failures before deployment. ## Core Features & Use Cases - TDD-based skill authoring: Run baseline pressure scenarios with subagents (RED), write the minimal skill addressing observed failures (GREEN), then close rationalization loopholes (REFACTOR). - Discovery optimization (CSO): Guidance for writing trigger-focused descriptions, keyword coverage, token-efficient structure, and correct cross-referencing between skills. - Testing methodology reference: The companion file testing-skills-with-subagents.md details pressure types, rationalization tables, red-flag lists, and meta-testing techniques. - Use Case: You want to add a new discipline-enforcing skill (e.g., verification-before-completion). Run a scenario where an agent is under time and sunk-cost pressure without the skill, record its exact rationalizations, write the skill to counter them, and re-test until the agent complies under maximum pressure. ## Quick Start Ask your agent to use the writing-skills skill to create a new skill for a recurring workflow, starting with a baseline pressure scenario before writing any SKILL.md content.
